Addiction and the : the role of in the cause and treatment of dependence Review Synthèse Denise M. Tomkins,*† Edward M. Sellers*†‡§¶

Abstract From *the Centre for and Mental RECENT SCIENTIFIC ADVANCES HAVE LED to a greater understanding of the neurobiologi- Health, Toronto, Ont.; cal processes that underlie drug abuse and addiction. These suggest that multiple the Departments of systems may play a key role in the development and expression of †Pharmacology, ‡Medicine drug dependence. These advances in our knowledge promise not only to help us and §, University identify the underlying cause of drug abuse and dependence, but also to aid the of Toronto, Toronto, Ont; development of effective treatment strategies. and ¶the Centre for Research in Women’s Health, Sunnybrook & Women’s he chemicals that humans abuse are structurally diverse and produce dif- College Health Sciences ferent behavioural effects in the user. Nevertheless, all share the common Centre, Toronto, Ont. T feature that they can modulate the brain that is fundamental to initiating and maintaining behaviours important for survival (e.g., eating, sexual This article has been peer reviewed. activity).1 Researchers first postulated that specific neural circuits within the brain were involved in the regulation of reward processes when early studies demon- CMAJ 2001;164(6):817-21 strated that rats would press a lever in order to obtain electrical of cer- tain areas of the brain, but not others. The medial bundle (MFB), which connects the (VTA) to the (NAcc), was the site first identified in this way (Fig. 1). Other neurotransmitter pathways pro- jecting from the VTA and the NAcc that innervate additional limbic (e.g., the ) and cortical areas of the brain, which are important for the expression of emotions, reactivity to conditioned cues, planning and judgement (Fig. 1), have also been implicated in reward. Although the MFB consists of that contain , noradrenaline and (5-HT), it is the projection that has been most closely implicated in reward. Thus, natural and artifical rewards (food, sex, of abuse) have been shown to activate this dopaminergic pathway, also known as the mesolimbic dopamine pathway, causing an increase in dopamine levels within the NAcc. From an evolutionary perspective, this brain reward circuit has ensured survival by giving priority to essential actions such as reproduction. Drugs of abuse are able to exert influence over the brain reward pathway either by directly influencing the action of dopamine within the system, or by altering the activity of other neurotransmitters that exert a modulatory influence over this mesolimbic dopaminergic pathway. γ-Aminobutyric acid (GABA), , seroton- ergic, cholinergic and noradrenergic neurotransmitter pathways have all been shown to interact at various points along the mesolimbic dopaminergic pathway and to modulate its activity. Some of the major elements in the brain reward circuit are illustrated in Fig. 2. Advances in our knowledge of the mechanisms by which various drugs of abuse al- ter the activity of this neuroanatomical system have helped our understanding of the neurobiological underpinnings of addiction and in the development of effective treat- ment strategies. To illustrate some of the advances in research and the application of this knowledge, a brief review of the literature is provided. This has been subdivided by drug class and presented in order of the prevalence of each drug’s use in Canada.

Nicotine is the main psychoactive constituent found in that has been shown to be responsible for its behavioural and physiological effects, which can lead

CMAJ • MAR. 20, 2001; 164 (6) 817

© 2001 Canadian Medical Association or its licensors Tomkins and Sellers

to addiction. Nicotine exerts its effects in the brain by act- nisms of action are reported to be via interactions with the ing on a specific type of receptor for the neurotransmitter noradrenergic and dopaminergic systems implicated in , known as the nicotinic receptor. Some nico- , however, the nature of its mechanism tinic receptors are located on the cell bodies of dopamine of action is poorly understood at the present time. Multiple neurons within the VTA, and activation of these receptors clinical trials have demonstrated that is an effec- increases the activity of these dopamine neurons, leading to tive aid in facilitating , although, in the an increase in dopamine release in the NAcc, which is majority of cases, the from smoking was not thought to mediate reward.2,3 Nicotinic receptors are also maintained in the long term.8 located on other neurotransmitter inputs to the VTA and further increase dopamine release by removing the in- hibitory influence that these other neurotransmitter inputs exert over the dopamine neurons (Fig. 2). The role of the Along with nicotine, alcohol is one of the most com- mesolimbic dopamine system in nicotine reward has been monly used psychoactive substances in Canada. Multiple clearly demonstrated in animals, because both the adminis- neurotransmitter systems play a role in mediating the be- tration of dopamine blockers and the destruction of the havioural effects of alcohol that have been linked to its mesolimbic dopamine pathway with selective abuse and dependence.9 This undoubtedly reflects the fact attenuate nicotine self-administration.4 that alcohol produces many pharmacological effects within Similarly, research in humans suggests a role for the brain. Nonetheless, the mesolimbic dopamine system dopaminergic processes in regulating the reinforcing ef- has been shown to play a role in the rewarding effects of al- fects of nicotine. In the limited number of studies con- cohol. Alcohol is similar to other abused substances in that ducted to date, administration of dopamine blockers has it increases NAcc dopamine release, and blocking the ef- been found to alter smoking behaviour. However, in the fects of dopamine reduces alcohol intake by animals.9 Fur- case of humans, dopamine blockade leads to a compen- thermore, innate differences in central dopaminergic neu- satory increase in, and not suppression of, smoking.4 Other neurotransmitter systems have also been implicated, in- GABA interneuron tonically suppresses dopamine cluding the opioid and cholinergic systems. Thus, adminis- cell firing, resulting in reduced VTA tration of either an opioid or reduced NAcc dopamine release. smoking behaviour in some, but not all, studies reported.5,6 , nicotine and alcohol can block the The most commonly used pharmacological approach for exerted by these neurons over the VTA dopamine cell bodies, treating nicotine dependence is the use of nicotine replace- resulting in increased VTA ment therapies, such as the nicotine patch, which reduce dopamine activity. craving by maintaining plasma nicotine levels.7 However, further advances in our understanding of the neurobiologi- cal factors underlying nicotine dependence have led to the introduction of new pharmacotherapies for its treatment. NAcc Dopamine cell body The nonnicotine medication, bupropion, which was origi- Activation results in the release of dopamine in the NAcc. nally developed as an , has recently been ap- Dopamine proved as an aid to smoking cessation. Its primary mecha- Released dopamine interacts Opioids, nicotine and alcohol can with postsynaptic dopamine stimulate the dopamine cell body receptors, resulting in reward. directly by interacting with specific receptors on its surface and/or indirectly by altering the activity of other neurotransmitter inputs projecting from distal brain areas. Prefrontal Planning, cortex judgement Frontal cortex recycles some of the released dopamine back into the nerve terminal. Nucleus accumbens and Reward Medial forebrain block reuptake of dopamine, bundle which accumulates in the synapse Ventral tegmental where it can further stimulate area dopamine receptors. Amphetamines also cause Emotions, Amygdala conditioned effects dopamine release. Lianne Friesen Fig. 2: Schematic diagram that represents the dopamine path- Lianne Friesen way projecting from the ventral tegmental area (VTA) to the Fig. 1: Schematic diagram of the that highlights nucleus accumbens (NAcc), indicating how substances of some of the main brain areas and neurotransmitter pathways abuse can alter the activity of this pathway to produce their implicated in reward processes. rewarding effects.

818 JAMC • 20 MARS 2001; 164 (6) Addiction and the brain

rotransmission have been linked to high levels of alcohol cated that subjects had a reduced to drink and, in drinking in selectively bred rodent lines.10 However, ani- cases where alcohol was consumed, subjects reported a mals will continue to self-administer alcohol if the dampening in the hedonic properties of alcohol. More re- mesolimbic dopamine pathway is destroyed using a selec- cent clinical trials, however, have suggested that compli- tive . This suggests that additional mechanisms ance issues are important in determining treatment efficacy are involved in regulating the rewarding effects of alcohol. with , and strategies for improving treatment Other neurotransmitter systems that have been implicated outcomes are the subject of current research.20 include the serotonergic, glutamatergic, GABAergic and is the drug of choice for the treatment of opioid systems. Thus, alcohol directly binds to and modu- in Europe. It is a structural analogue of the neu- lates the activity of various specific receptors of these neu- rotransmitter GABA, which is reported to exert its effects rotransmitter systems (e.g., 5-HT3, GABAA and N-methyl- mainly via an interaction with the NMDA receptor. As D-aspartate [NMDA] receptors) that are located within the noted earlier, the NMDA receptor is one of the specific re- brain reward pathway and can indirectly modulate ceptors located within the brain reward pathway to which mesolimbic dopamine activity via feedback mechanisms, for alcohol directly binds, and it can indirectly modulate example, by increasing VTA activity through decreasing mesolimbic dopamine activity. This is relevant to alco- the suppressant influence that GABAergic inputs exert over holism because hyperexcitability in the NMDA system has them (Fig. 2).11 been shown to occur following long-term alcohol use and Altered central dopamine function has also been impli- has been linked with the expression of withdrawal on cessa- cated as influencing the propensity for alcohol consump- tion of drinking. Clinical testing of the efficacy of acam- tion in humans, at least in some populations.12 Human ge- prosate in the treatment of alcoholism has been exten- netic studies suggest that an association exists between sive.21–25 The outcomes of these studies have consistently alcoholism and both the dopamine D2 receptor and the demonstrated that the use of acamprosate significantly im- dopamine transporter. This is supported by brain imaging proves abstinence rates and reduces the number of . studies that have reported alterations in both D2 receptor In contrast to naltrexone, compliance issues do not appear and dopamine transporter densities in the of alco- to be as much of a concern with acamprosate. A large mul- holics.13–15 These findings could have important implica- ticentre trial with acamprosate is currently underway in the tions for treatment, because recent work suggests that in- United States.26 creased density of D2 receptors may be a predictor of In contrast to many other abused substances, the effects vulnerability to in alcohol-dependent patients.16 As of alcohol on the brain reward system are quite complex, as highlighted by animal studies, other neurotransmitter sys- highlighted earlier, with many neurotransmitter systems tems have similarly been implicated as underlying the being implicated. This presents unique challenges for the propensity to consume alcohol in humans, and the newer development of effective pharmacotherapies for the treat- pharmacotherapies that have been approved for the treat- ment of alcohol abuse and dependence. It is likely that ment of mediate their effects via some pharmacological interventions directed at altering the in- of these systems. fluence of one neurotransmitter system on the rewarding The most common treatment strategies for alcoholism effects of alcohol may have limited utility because of com- that are currently employed are psychosocial interventions pensatory changes in the other regulatory neurotransmitter and self-help groups (e.g., ), because systems. This is supported by the experiences reported with the use of pharmacotherapies has been limited until re- various pharmacotherapeutic agents used singly in the clin- cently by the lack of effective therapeutic agents. In the ical setting, in which small but significant improvements in past, pharmacological agents were used mainly to alleviate alcohol-drinking indices were noted.17 Thus, a broader the symptoms of acute withdrawal (e.g., benzodiazepines pharmacotherapeutic approach using a combination of and β-blockers to reduce ), to prevent the complica- drugs with different mechanisms may prove to be more ef- tions of withdrawal (e.g., anticonvulsant agents to prevent fective. A recent study supports the notion that such a seizures) or to provide an aversive experience when alcohol treatment strategy can improve patient outcomes27 and fur- was ingested (e.g., ).17 More recently, 2 drugs ther improve compliance by reducing the reported fre- have been approved for clinical use that are aimed specifi- quency of side effects. Although there has been limited re- cally at reducing alcohol consumption. Naltrexone, an opi- search in this area, these initial findings support the need oid receptor antagonist, reliably reduces alcohol intake in for further work. preclinical animal studies, which is consistent with the im- portance of the opioid system in modulating the rewarding effects of alcohol. Within North America, naltrexone is the most widely used pharmacotherapy for the treatment of al- drugs, such as cocaine and amphetamines, are coholism. Initial clinical trials reported that naltrexone sig- substances that typically cause heightened feelings of well- nificantly improved abstinence rates and reduced relapse being and , and an increased state of arousal. As rates in recently abstinent alcoholics.18,19 Self-reports indi- with most drugs abused by humans, the psychostimulants

CMAJ • MAR. 20, 2001; 164 (6) 819 Tomkins and Sellers

are similarly self-administered by animals, and most of the mediate their reinforcing effects by modulating the activity information regarding their mechanism of action has been of the , although not directly.39 The derived from animal research. From a biochemical perspec- enhance NAcc dopamine release by increasing the tive, the major mechanism by which the amphetamines and activity of VTA dopamine neurons. It is postulated that this cocaine potentiate the actions of dopamine within the is achieved via activation of mu-opioid receptors located on mesolimbic system is by inhibiting dopamine reuptake into GABA neurons within the VTA, which play an important the nerve terminals via the dopamine transporter.1,2 The role in regulating the activity of VTA dopamine neurons. importance of these pharmacodynamic effects within the Opiates also have dopamine-independent effects within the brain with respect to drug-taking behaviour has been NAcc, which play an important role in reward.40 Evi- clearly demonstrated in animal models. For example, psy- dence supporting these proposed mechanisms of action in- chostimulants are self-administered to a lesser extent in an- cludes the observation that and self- imals in which lesions of the mesolimbic dopamine path- administration can be modified by specifically blocking the way have been produced by the application of a selective actions of GABA in the VTA and opioid receptors in the neurotoxin or when pretreated with NAcc. blockers.2 Furthermore, mice that have been bred to have Functional in opiate addicts confirms that no dopamine transporter (known as DAT knockout mice) heroin administration and heroin-related cues activate the are unresponsive to the stimulant effects of same neural systems identified in preclinical animal and cocaine.28 These mice also fail to show elevations in ex- studies,41 providing evidence that these systems modulate tracellular dopamine levels following psychostimulant ad- the direct and conditioned pharmacological effects of the ministration, clearly demonstrating the importance of the drug. The latter effects are an important element of drug dopamine transporter in mediating the effects of the psy- craving. In addition, pharmacological challenges with drugs chostimulants. interacting with dopamine receptors have demonstrated Although research in humans is more limited, the find- that the activity of brain dopaminergic systems has been al- ings confirm that similar pharmacodynamic effects occur tered by long-term opiate use in heroin addicts,42 and these following psychostimulant administration in humans.29 changes are proposed to be related to the neural processes Imaging studies have demonstrated that subcortical re- leading to abuse and dependence. The treatment strategies gions within the extended amygdala are activated both fol- for opiate dependence using pharmacological agents have lowing cocaine infusion30 and in response to cue-induced involved 3 different approaches. The most common strat- cocaine craving.31 Research using positron emission to- egy employed is a approach by substituting mography imaging techniques has shown that the rein- less harmful long-acting, orally active opioid for forcing effects of psychostimulants are correlated with in- the more harmful abused substance (e.g., and creases in dopamine concentrations in limbic regions due maintenance programs).43,44 Another ap- to blockade of the dopamine transporter and occupancy of proach has been to alleviate symptoms resulting from acute 32,33 dopamine D2 receptors. Chronic psychostimulant use withdrawal (e.g., using clonidine), which has proved suc- has been shown to cause long-term changes in the func- cessful for facilitating detoxification.45 A final treatment 34,35 tion of the dopamine transporter and in D2 receptor strategy has been to block the pharmacological conse- levels, which may contribute to further psychostimulant quences of further opiate use following detoxification by abuse. Despite the extensive evidence for a primary role administering an opioid blocker (e.g., naltrexone). How- for the dopaminergic system in the stimulus–reward prop- ever, clinical trials suggest that this approach is mainly ac- erties of cocaine and amphetamines, clinical trials of ceptable to relatively opiate-free addicts who are highly agents that modify dopamine release or block its receptors motivated for change.46 have been disappointing.36 More recently, however, a criti- cal role for selective dopamine receptor subtypes (e.g., D1 Conclusion receptors) in the self-administration and conditioned re- sponses to cocaine suggests that agents directed at this re- The identification of the underlying biological mecha- ceptor may be effective.37,38 nisms of drug dependence has lead to targeted drug devel- opment. In the case of alcohol, both the opioid and Opiates NMDA systems have been targeted and have resulted in marketed products, namely, naltrexone and acamprosate Opiate drugs, such as morphine and codeine, are com- respectively. On the other hand, the discovery that bupro- monly used in the clinical setting for pain relief. However, pion is effective in treating tobacco dependence was clinical this class of psychoactive agent can also elicit intense eu- serendipity. This will, however, probably lead to the recog- phoric effects followed by feelings of well-being in the user nition of a new contributing neurochemical basis for nico- when taken in high doses, which can lead to their abuse and tine dependence. Perhaps it is most surprising that there ultimately may result in addiction. Animal research sug- are no effective pharmacological interventions for cocaine gests that, as with the psychostimulants, opiates appear to or stimulant abuse despite our extensive knowledge about

820 JAMC • 20 MARS 2001; 164 (6) Addiction and the brain

how they exert their effects in the brain. This may be a re- 22. Whitworth AB, Fisher F, Lesch OM, Nimmerrichter A, Oberbauer H, Platz T, et al. Comparison of acamprosate and placebo in long-term treatment of flection of the fact that these substances cause long-term alcohol dependence. Lancet 1996;347:1438-42. and often permanent changes in the brain. Current re- 23. Pelc I, Verbanck P, Le Bon O, Gavrilovic M, Lion K, Lehert P. Efficacy and safety of acamprosate in the treatment of detoxified alcohol-dependent pa- search efforts are now also focusing on exploring the neu- tients. A 90-day placebo-controlled dose-finding study. Br J Psychiatry 1997; robiological factors that may underlie relapse into drug- 171:73-7. 24. Poldrugo F. Acamprosate treatment in a long term community-based alcohol taking behaviour. This is an important issue because a rehabilitation program. Addiction 1997;92:1537-46. successful treatment strategy should be effective in prevent- 25. Geerlings PJ, Ansoms C, van den Brink W. Acamprosate and prevention of relapse in alcoholics. Eur Addict Res 1997;3:129-37. ing relapse, which occurs with high prevalence in most 26. Johnson BA, Ait-Daoud N. Neuropharmalogical treatments for alcoholism: drug- and alcohol-abusing populations. scientific basis and clinical findings. 2000;149:327-44. 27. Johnson BA, Ait-Daoud N, Prihoda TJ. Combining ondansetron and naltrex- one effectively treats biologically predisposed alcoholics: from hypothesis to Competing interests: None declared. preliminary clinical evidence. Alcohol Clin Exp Res 2000;24:737-42. 28. Giros B, Jaber M, Jones SR, Wightman RM, Caron MG. Hyperlocomotion Contributors: Dr. Tomkins is the principal author of this review. Dr. Sellers also and indifference to cocaine and amphetamine in mice lacking the dopamine participated in the writing and drafted the original figures. transporter. Nature 1996;379:606-12. 29. Baxter LR Jr, Schwartz JM, Phelps ME, Mazziotta JC, Barrio RA, Engel J, et al. Localization of neurochemical effects of cocaine and other stimulants in References the human brain. J Clin Psychiatry 1988;49:23-6. 30. Breiter HC, Rosen BR. Functional magnetic resonance imaging of brain re- 1. Koob GF, Sanna PP, Bloom FE. of addiction. 1998;21: ward circuitry in the human. Ann N Y Acad Sci 1999; 877:523-47. 467-76. 31. Childress AR, Mozley PD, McElgin W, Fitzgerald J, Reivich M, O’Brien CP. 2. Corrigall WA. Understanding brain mechanisms in nicotine . Limbic activation during cue-induced cocaine craving. Am J Psychiatry Br J Addict 1991;86:507-10. 1999;156:11-8. 3. Picciotto MR. Common aspects of the action of nicotine and other drugs of 32. Volkow ND, Wang GS, Fowler JS, Logan J, Gatley SJ, Wong C, et al. Rein- abuse. Drug Alcohol Depend 1998;51:165-72. forcing effects of psychostimulants in humans are associated with increases in 4. Rose JE, Corrigall WA. Nicotine self-administration in animals and humans: brain dopamine and occupancy of D(2) receptors. J Pharmacol Exp Ther similarities and differences. Psychopharmacology 1997;130:28-40. 1999;291:409-15. 5. Perkins KA, Sanders M, Fonte C, Wilson AS, White W, Stiller R, et al. Ef- 33. Volkow ND, Fowler JS, Wang GJ. Imaging studies on the role of dopamine fects of central and peripheral nicotinic blockade on human nicotine discrimi- in cocaine reinforcement and addiction in humans. J Psychopharmacol 1999; nation. Psychopharmacology 1999;142:158-64. 13:337-45. 6. Brauer LH, Behm FM, Westman EC, Patel P, Rose JE. Naltrexone blockade 34. Little KY, Carroll FI, Butts JD. Striatal [125I]RTI-55 binding sites in cocaine- of nicotine effects in cigarette smokers. Psychopharmacology 1999;143:339-46. abusing humans. Prog Neuropsychopharmacol Biol Psychiatry 1998;22:455-66. 7. Jorenby DE. New developments in approaches to smoking cessation. Curr 35. Little KY, McLaughlin DP, Zhang L, McFinton PR, Dalack GW, Cook EH, Opin Pulm Med 1998;4:103-6. et al. Brain dopamine transporter messenger RNA and binding sites in co- 8. Hughes JR, Goldstein MG, Hurt RD, Shiffman S. Recent advances in the caine users: a postmortem study. Arch Gen Psychiatry 1998;55:793-9. pharmacotherapy of smoking. JAMA 1999;281:72-6. 36. Klein M. Research issues related to development of medications for treatment 9. Koob GF, Weiss F. of cocaine and ethanol dependence. of cocaine addiction. Ann N Y Acad Sci 1998;844:75-91. In: Galanter M, editor. Recent developments in alcoholism. vol 10. New York: 37. Haney M, Collins ED, Ward AS, Foltin RW, Fischman MW. Effect of a se- Plenum Press; 1992. p. 201-33. lective dopamine D1 (ABT-431) on smoked cocaine self-administra- 10. Li TK. Pharmacogenetics of responses to alcohol and that influence al- tion in humans. Psychopharmacology 1999;143:102-10. cohol drinking. J Stud Alcohol 2000;61:5-12. 38. Romach MK, Glue P, Kampman K, Kaplan HL, Somer GR, Poole S, et al. 11. Grant KA. Emerging neurochemical concepts in the actions of ethanol at Attenuation of the euphoric effects of cocaine by the dopamine D1/D5 antag- ligand-gated ion channels. Behav Pharmacol 1994;5:383-404. onist . Arch Gen Psychiatry 1999;56:1101-6. 12. Cowen MS, Lawrence AJ. The role of opioid-dopamine interactions in the 39. Shippenberg TS, Elmer GI. The neurobiology of opiate reinforcement. Crit induction and maintenance of ethanol consumption. Prog Neuropsychopharma- Rev Neurobiol 1998;12:267-303. col Biol Psychiatry 1999;23:1171-212. 40. Koob GF, Bloom FE. Cellular and molecular mechanisms of drug depen- 13. Repo E, Kuikka JT, Bergstron KA, Karhu J, Hiltunen J, Tiihonen J. dence. Science 1988;242:715-23. Dopamine transporter and D2-receptor density in late-onset alcoholism. Psy- 41. Sell LA, Morris J, Bearn J, Frackowiak RS, Friston KJ, Dolan RJ. Activation chopharmacology 1999;147:314-8. of reward circuitry in human opiate addicts. Eur J Neurosci 1999;11:1042-8. 14. Volkow ND, Wang GJ, Fowler JS, Logan J, Hitzemann R, Ding YS, et al. 42. Casas M, Guardia J, Prat G, Trujols J. The apomorphine test in heroin ad- Decreases in dopamine receptors but not in dopamine transporters in alco- dicts. Addiction 1995;90:831-5. holics. Alcohol Clin Exp Res 1996;20:1594-8. 43. Warner EA, Kosten TR, O’Connor PG. Pharmacotherapy for opioid and co- 15. Tupala E, Hall H, Sarkioja T, Rasanen P, Tiihonen J. Dopamine-transporter caine abuse. Med Clin North Am 1997;81:909-25. denisty in nucleus accumbens of type-1 alcoholics. Lancet 2000;355:380. 44. Effective medical treatment of opiate addiction. National Consensus Devel- 16. Guardia J, Catafau AM, Battle F, Martin JC, Segura L, Gonzalvo B, et al. opment Panel on Effective Medical Treatment of Opiate Addiction. JAMA Striatal dopaminergic D(2) receptor density measured by [(123)I]iodobenza- 1998;280:1936-43. mide SPECT in the prediction of treatment outcome of alcohol-dependent 45. Gerra G, Marcato A, Caccavari R, Fontanesi B, Delsignore R, Fertonani G, patients. Am J Psychiatry 2000;157:127-9. et al. Clonidine and opiate receptor antagonists in the treatment of heroin ad- 17. Gatch MB, Lal H. Pharmacological treatment of alcoholism. Prog Neuropsy- diction. J Subst Abuse Treat 1995;12:35-41. chopharmacol Biol Psychiatry 1998;22:917-44. 46. Clinical evaluation of naltrexone treatment of opiate-dependent individuals. 18. Volpicelli JR, Alterman AI, Hayasgida M, O’Brien CP. Naltrexone in the Report of the National Research Council Committee on Clinical Evaluation treatment of alcohol dependence. Arch Gen Psychiatry 1992;49:876-80. of Antagonists. Arch Gen Psychiatry 1978;35:335-40. 19. O’Malley SS, Jaffe AJ, Chang G, Schottenfeld RS, Meyer RE, Rounsaville B. Naltrexone and coping skills therapy for alcohol dependence. Arch Gen Psychi- atry 1992;49:881-7. 20. Swift RM. Drug therapy for alcohol dependence. N Engl J Med 1999;340: 1482-90. Reprint requests to: Dr. Denise M. Tomkins, Centre for Addiction 21. Lhuintre JP, Moore N, Tran G, Steru L, Langrenon S, Daoust, et al. Acam- prosate appears to decrease alcohol intake in weaned alcoholics. Alcohol Alcohol and , 33 Russell St., Toronto ON M5S 2S1; 1990;25:613-22. fax 416 595-6922; [email protected]

CMAJ • MAR. 20, 2001; 164 (6) 821